de Recherche sur l'Environnement Marin (CEFREM), CNRS/Université de Perpignan, 52 Avenue de Villeneuve, 66860 Perpignan Cedex, France. e-mail: ludwig@univ-perp.fr 1. Mackenzie, F. T. & Garrels, R. M. Am. J. Sci. 264, 507–525 (1966). 2. Smith, S. V. & Hollibaugh, G. T. Rev. Geophys. 31, 75–89 (1993). 3. Williams, P. M. & Druffel, E. R. M. Nature 330, 246–248 (1987). 4. Raymond, P. A. & Bauer, J. E. Nature 409, 497–500 (2001). 5. Hedges, J. I., Keil, R. G. & Benner, R. Org. Geochem. 27, 195–212 (1997). 6. Ludwig, W., Probst, J.-L. & Kempe, S. Glob. Biogeochem. Cycles 10, 23–41 (1996). 7. Kao, S. & Liu, K. Limnol. Oceanogr. 41, 1749–1757 (1996). 8. McClain, M. E., Richey, J. E. & Brandes, J. A. Glob. Biogeochem. Cycles 11, 295–311 (1997). number of legs, humans and quadrupeds change from walk to run or trot at the same Froude number, close to 0.5 (ref. 3). Even gibbons 8 change from merely swinging to swinging with aerial phases at Froude num- bers ranging from 0.3 to 0.6. Humans of short height, such as children 9,10 , patients suffering from early-onset growth-hormone deficiency 11 and pygmies 12 , optimally walk at speeds that correspond dynamically to that for adults, that is, with a Froude number of 0.25. Here, walking speed is optimal when the recovery of energy by the body — by exchanging potential and kinetic energies — is maximal 7 . What all of this means is that, within the same gravitational environment (such as on the Earth), the smaller the body, the lower the ‘equivalent’ speed of move- ment, which is proportional to the square root of the leg length. But the power of the Froude number for predicting equivalent walking speeds is not confined to the Earth. Within the same species and for a given body size, dynamic news and views Presynaptic active zones are the specialized sites from which nerves release neurotransmitter. They are characterized by the presence of calcium channels, synaptic vesicles containing neurotransmitter, and aggregates of protein that make up what is known as ‘active-zone material’. This material was described by S. L. Palay almost 50 years ago (J. Biophys. Biochem. Cytol. 2, 193–202; 1956), but its precise structure and function have remained an enigma. Things may be about to change, however, thanks to the publication elsewhere in this issue of stunning three- dimensional reconstructions of the frog neuromuscular junction (Nature 409, 479–484; 2001). To study this junction between nerve and muscle, M. L. Harlow and colleagues used a technique known as electron microscope tomography, which gives greater spatial resolution than conventional transmission electron microscopy. They describe three distinct elongate structures — which they call ribs, beams and pegs — that make up the first 15 nanometres of active-zone material as one looks into the neuron. The top image shows this view, with the active-zone material in yellow and the synaptic vesicles in blue. Beams — which make up the vertical band in the centre of this image — are arranged along the long axis of the elongate active zone, and form the backbone of the structure. Ribs lie perpendicular to this long axis, and link the central backbone to the double row of synaptic vesicles that straddle it. Pegs (not apparent in these images) tether each rib to the plasma membrane in a pattern that matches the distribution of molecules that are thought to be calcium channels. The lower image is a side-on view, with the presynaptic plasma membrane in grey. The structure is perhaps best appreciated in the movies published on Nature’s web site (http://www.nature.com/nature/). As a whole, the active-zone material appears to provide a scaffold by which synaptic vesicles are localized to the specialized presynaptic membrane. But the proteins that make up the structure probably have more specific functions. The intracellular domains of calcium channels may form all or part of the pegs, and ribs may be composed of the host of proteins that together mediate docking and fusion of vesicles with the presynaptic membrane. We eagerly await the formal identification of the rib, beam and peg proteins. Lesley Anson Neurobiology Activity at the active zone Biomechanics Walking on other planets Alberto E. Minetti A nineteenth-century equation used for building model ships allows us to compare the motion of animals of different sizes and gaits. It may also give us an idea of how we would move on different planets. W hat do a gibbon swinging by its arms from tree to tree, a sailing steamship, and a walking human have in com- mon? The answer lies in a simple concept introduced in the nineteenth century by nautical engineer William Froude to help him to produce model ships that maintained the same propulsion dynamics as full-size vessels. His ‘Froude number’ is a dimension- less variable that was brought into biology by D’Arcy Thompson 1 and popularized by McNeill Alexander 2,3 in the study of the energetics and mechanics of animal loco- motion. This number allows one to compare the motion of species with different numbers of legs and gaits, and to investigate the effects of different body sizes on the mechanics of movement. As suggested by various earlier papers 4–6 , and borne out by new work described by Cavagna and colleagues in the Journal of Physiology 7 , the Froude number is also a reliable rule-of-thumb for predict- ing walking speeds on planets with gravities different from that of the Earth. In walking humans or swinging gib- bons, changes in the vertical position of the body’s centre of mass during movement affect the gravitational potential energy of the body, and are accompanied by opposite changes in the kinetic energy needed to drive movement. The result is a pendulum- like mode of movement that saves mech- anical energy. Roughly the same principles apply to ships, but here the potential energy is related to the size of the wave generated by the ship. One of the theories underlying studies of movement — dynamic similarity 3 — basically states that geometrically similar bodies that rely on pendulum-like mech- anics of movement will have similar gait dynamics if the Froude number remains the same. This number (Fr) is given by Frǃv 2 /(gǂl), where v is the speed of move- ment (in m s ǁ1 ), g is acceleration due to gravity (in m s ǁ2 ) and l is a characteristic length (such as leg length, in metres). The Froude number is directly proportional to the ratio between the kinetic energy and the gravitational potential energy needed during movement. Dynamic similarity implies that, for example, despite differences in body size and NATURE | VOL 409 | 25 JANUARY 2001 | www.nature.com 467 M. L. HARLOW ET AL. © 2001 Macmillan Magazines Ltd